Map-less and localization-less lane following method for autonomous driving of autonomous driving vehicles on highway

Patent metadata

Field	Value
Publication number	US-11260880-B2
Application number	US-201816067556-A
Country	US
Kind code	B2
Filing date	Apr 18, 2018
Priority date	Apr 18, 2018
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

In one embodiment, instead of using map data, a relative coordinate system is utilized to assist perception of the driving environment surrounding an ADV for some driving situations. One of such driving situations is driving on a highway. Typically, a highway has fewer intersections and exits. The relative coordinate system is utilized based on the relative lane configuration and relative obstacle information to control the ADV to simply follow the lane and avoid potential collision with any obstacles discovered within the road, without having to use map data. Once the relative lane configuration and obstacle information have been determined, regular path and speed planning and optimization can be performed to generate a trajectory to drive the ADV. Such a perception system is referred to as a relative perception system based on a relative coordinate system.

First claim

What is claimed is:

1. A computer-implemented method for generating a trajectory for operating an autonomous driving vehicle to follow lanes, the method comprising:
determining a driving environment surrounding an autonomous driving vehicle (ADV) driving on a lane based on sensor data obtained from a plurality of sensors;
determining a lane configuration of the lane relative to a current location of the ADV based on perception data of the driving environment, without using map data of a map associated with the lane;
determining obstacle information of an obstacle relative to the ADV, including a relative position of the obstacle relative to the current location of the ADV;
generating a local view frame based on the lane configuration and the obstacle information, the local view frame describing the lane configuration and the obstacle from a view point of the ADV without map information;
generating a trajectory for the ADV to follow the lane for a next driving cycle based on the local view frame;
generating a reference line that represents a center line within the lane;
generating a station-lateral (SL) map based on the lane configuration and the obstacle information, wherein the SL map includes information describing a lateral position of the obstacle relative to the reference line and a vertical position relative to the current location of the ADV, and wherein the SL map is utilized to optimize a shape of the trajectory; and
controlling the ADV based on the trajectory to follow the lane.

2. The method of claim 1 , further comprising, prior to determining the lane configuration of the lane, determining that the ADV is driving on a highway based on the perception data of the perceived driving environment, wherein the lane configuration is determined in response to determining that the ADV is driving on a highway.

3. The method of claim 1 , further comprising, for each of a plurality of driving cycles,
iteratively performing determining the driving environment, determining the lane configuration, determining the obstacle information, and generating the local view frame for the corresponding driving cycle wherein the local view frames associated with the driving cycles are utilized to create the trajectory.

4. The method of claim 1 , wherein the lane is one of a plurality of lanes of a road in which the ADV is driving, wherein generating a trajectory to drive the ADV comprises:
generating a plurality of local view frames, one for each of the plurality of lanes based on the lane configuration and the obstacle information;
generating a plurality of paths, one for each of the local view frames; and
selecting one of the paths based on a path cost associated with each of the paths using a predetermined cost function, wherein the selected path is utilized to generate the trajectory.

5. The method of claim 1 , wherein determining lane configuration of the lane relative to the ADV comprises:
measuring a first distance between the current location of the ADV and a first edge of the lane based on a first image of the first edge;
measuring a second distance between the current location of the ADV and a second edge of the lane based on a second image of the second edge; and
calculating a lane width of the lane based on the first distance and the second distance.

6. The method of claim 1 , wherein the reference line is generated based on the lane width of the lane and based on the lane configuration without the map data.

7. The method of claim 1 , further comprising generating a station-time (ST) graph based on the lane configuration and the obstacle information, wherein the ST graph includes information describing a relative position of the obstacle relative to the reference line at different points in time, and wherein the ST graph is utilized to optimize a speed of the ADV at different points in time along the trajectory.

8. The method of claim 1 , wherein the current location of the ADV is measured based on a position of a center of a rear axle of the ADV.

9. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations, the operations comprising:
determining a driving environment surrounding an autonomous driving vehicle (ADV) driving on a lane based on sensor data obtained from a plurality of sensors of the ADV, including perceiving an obstacle;
determining a lane configuration of the lane relative to a current location of the ADV based on perception data of the driving environment, without using map data of a map associated with the lane;
determining obstacle information of an obstacle relative to the ADV, including a relative position of the obstacle relative to the current location of the ADV;
generating a local view frame based on the lane configuration and the obstacle information of the obstacle, the local view frame describing the lane configuration and the obstacle from a view point of the ADV without map information;
generating a trajectory to follow the lane for a next driving cycle based on the local view frame;
generating a reference line that represents a center line within the lane;
generating a station-time (ST) graph based on the lane configuration and the obstacle information, wherein the ST graph includes information describing a relative position of the obstacle relative to the reference line at different points in time, and wherein the ST graph is utilized to optimize a speed of the ADV at different points in time along the trajectory; and
controlling the ADV based on the trajectory to follow the lane.

10. The machine-readable medium of claim 9 , wherein the operations further comprise, prior to determining the lane configuration of the lane, determining that the ADV is driving on a highway based on the perception data of the perceived driving environment, wherein the lane configuration is determined in response to determining that the ADV is driving on a highway.

11. The machine-readable medium of claim 9 , wherein the operations further comprise, for each of a plurality of driving cycles,
iteratively performing determining the driving environment, determining the lane configuration, determining the obstacle information, and generating the local view frame for the corresponding driving cycle wherein the local view frames associated with the driving cycles are utilized to create the trajectory.

12. The machine-readable medium of claim 9 , wherein the lane is one of a plurality of lanes of a road in which the ADV is driving, wherein generating a trajectory to drive the ADV comprises:
generating a plurality of local view frames, one for each of the plurality of lanes based on the lane configuration and the obstacle information;
generating a plurality of paths, one for each of the local view frames; and
selecting one of the paths based on a path cost associated with each of the paths using a predetermined cost function, wherein the selected path is utilized to generate the trajectory.

13. The machine-readable medium of claim 9 , wherein determining lane configuration of the lane relative to the ADV comprises:
measuring a first distance between the current location of the ADV and a first edge of the lane based on a first image of the first edge;
measuring a second distance between the current location of the ADV and a second edge of the lane based on a second image of the second edge; and
calculating a lane width of the lane based on the first distance and the second distance.

14. The machine-readable medium of claim 9 , wherein the reference line is generated based on the lane width of the lane and base

Assignees

• Baidu Usa Llc
• Baidu Com Times Tech Beijing Co Ltd

Inventors

• Zhu Fan
• Kong Qi
• Pan Yuchang
• Jiang Feiyi
• Xu Xin
• Fu Xiaoxin
• Xia Zhongpu
• Zhao Chunming
• Zhang Liangliang
• Zhu Weicheng
• Zhuang Li
• Fan Haoyang
• Jiang Hui
• Tao Jiaming

Classifications

• B60W60/00276 Primary

  for two or more other traffic participants · CPC title

• B60W2552/53

  Road markings, e.g. lane marker or crosswalk · CPC title

• B60W2552/05

  Type of road, e.g. motorways, local streets, paved or unpaved roads · CPC title

• G08G1/167

  Driving aids for lane monitoring, lane changing, e.g. blind spot detection · CPC title

• B60W30/143

  Speed control (B60W30/16 takes precedence) · CPC title